Audience Level:
All
Institutional Level:
Higher Ed
Special Session:
Research
Abstract:
Data has been called “the new oil” and online education generates substantial data reserves. In this presentation, we’ll discuss applied research projects using data science for student success, including predictive modeling, automated writing feedback, and course design analysis. We will also discuss ethical considerations and next steps in the field.
Extended Abstract:
- Introduction to Data Science & Learning Analytics
- MOOCs, online courses,
- Distinguish academic analytics and learning analytics
- Predictions about importance and use of analytics in higher ed
- Research Examples
- Stealth Assessment of Social and Emotional Skills using Learning Analytics – current work in progress (results September 2019) to evaluate the relationship between SE skills and online activity (LMS and e-Textbook). Will investigate “learning patterns” using sequential data mining instead of aggregate count analysis, and also use deep learning and advanced techniques.
- Early Course Intervention Analytics – discuss projects underway at UMBC to predict student performance early in the term and make interventions to improve their outcomes. Using multiple solutions and systems (BB Predict, EAB, Civitas) and can compare experiences and adoption rates.
- Automated Writing Evaluation – discuss how “robograders” actually work in practice to simulate faculty decisions and research conducted upon accuracy before deployments are made. Evaluate features and rubrics used as well as deployments that supplement (not replace) human graers.
- Ethical Considerations – discuss current debates around student consent, institutional obligations to act, and frameworks created to enable productive conversations around ethics (e.g. Delicate framework)
- Next Steps & Directions
- Theory: Connection of learning analytics with learning constructs
- Practice: Publicly available tools (e.g. OpenSimon) and role of specialists
Conference Track:
Learner Services and Support
Session Type:
Education Session
Intended Audience:
All Attendees